Policy Evaluation

2 articles

When the parallel-trends test fails on one lead, what's left?

Replicating Wang et al. (2026) TWFE-DiD on SNAP BBCE adoption from IPUMS ACS, with Goodman-Bacon decomposition and leave-one-out diagnostics.

May 2026 · Methodology

A 21 percent policy effect that wasn’t: testing for Medicaid Unwinding contamination in DiD designs

A three-check diagnostic protocol for state Medicaid difference-in-differences designs whose post-treatment window touches April 2023, applied to a branding-and-expansion case with paired data release.

May 2026 · Methodology